For HTML parsing, you have the following possibilities:

QXML
QDOM
perl
khtml

Obviously, QXML and QDOM need xml-compliant html pages, and the least html pages are xml-compliant. Perl is not the scope of this site. So, this tutorial choses the khtml approach.

First step

As we remember from http://developernew.kde.org/Development/Tutorials/Programming_Tutorial_KDE_4/How_to_write_an_HTML_parser, biggest thing is to be able to parse non-XML-conform syntax. It works with the following program.

tags.cpp

include <kapplication.h>
include <kaboutdata.h>
include <kcmdlineargs.h>
include <dom/html_document.h>

int main (int argc, char *argv[])
{

       KAboutData aboutData( "test", "test",
       "1.0", "test", KAboutData::License_GPL,
       "(c) 2006" );
       KCmdLineArgs::init( argc, argv, &aboutData );
       KApplication khello;

       DOM::HTMLDocument doc;
       DOM::DOMString tag("*");
       DOM::DOMString uri("<html><body><a href=\"http://www.kde.org/\"></a><a href=\"/index.php\" nowrap>Log in</a><a href=\"http://www.gmx.de\"></a></body></html>");

       doc.loadXML(uri);
       kdDebug() << "Does this doc have child elements ? " << doc.hasChildNodes() << endl;
       for (int i=0; i<doc.getElementsByTagName(tag).length(); i++) kdDebug() << doc.getElementsByTagName(tag).item(i).nodeName().string() << endl;
       kdDebug() << "Size of your doc " << sizeof(doc.firstChild()) << endl;
       kdDebug() << doc.isHTMLDocument() << endl;
       kdDebug() << doc.toString().string() << endl;

}

Compile it like this:

gcc -I/usr/lib/qt3/include -I/opt/kde3/include \
-L/opt/kde3/lib -lkdeui -lkhtml -o tags tags.cpp

Second

include <kapplication.h>
include <kaboutdata.h>
include <kcmdlineargs.h>
include <dom/html_document.h>
include <dom/html_element.h>
include <dom/dom_node.h>

int main (int argc, char *argv[])
{

       KAboutData aboutData( "test", "test",
       "1.0", "test", KAboutData::License_GPL,
       "(c) 2006" );
       KCmdLineArgs::init( argc, argv, &aboutData );
       KApplication khello;

       DOM::HTMLDocument doc;
       DOM::DOMString tag("*");
       DOM::DOMString uri("<html><body><a href=\"http://www.kde.org/\">fat</a><a href=\"/index.php\" nowrap>Log in</a><a href=\"http://www.gmx.de\"></a></body></html>");

       doc.loadXML(uri);
       kdDebug() << "Here's a list of the document elements" << endl;
       for (int i=0; i<doc.getElementsByTagName(tag).length(); i++) kdDebug() << doc.getElementsByTagName(tag).item(i).nodeName().string() << endl;
      
       DOM::HTMLDocument doc2;
       DOM::DOMString uri2("<html><body>this is htmlfat</body></html>");
       doc2.loadXML(uri2);
       kdDebug() << "This is the in-memory html:" << endl;
       kdDebug() << doc.toString().string() << endl;
       doc.body().insertBefore(doc.body().firstChild().firstChild(),doc.body().firstChild());
       kdDebug() << "Moving around nodes" << endl;
       kdDebug() << doc.toString().string() << endl;

}